# On Coverage and Capacity for Disaster Area Wireless Networks Using Mobile Relays

- 770 Downloads
- 4 Citations

## Abstract

Public safety organizations increasingly rely on wireless technology to provide effective communications during emergency and disaster response operations. This paper presents a comprehensive study on dynamic placement of relay nodes (RNs) in a disaster area wireless network. It is based on our prior work of mobility model that characterizes the spatial movement of the first responders as mobile nodes (MNs) during their operations. We first investigate the COverage-oriented Relay Placement (CORP) problem that is to maximize the total number of MNs connected with the relays. Considering the network throughput, we then study the CApacity-oriented Relay Placement (CARP) problem that is to maximize the aggregated data rate of all MNs. For both coverage and capacity studies, we provide each the optimal and the greedy algorithms with computational complexity analysis. Furthermore, simulation results are presented to compare the performance between the greedy and the optimal solutions for the CORP and CARP problems, respectively. It is shown that the greedy algorithms can achieve near optimal performance but at significantly lower computational complexity.

## Keywords

Bipartite Graph Relay Node Mobility Model Disaster Area Binary Integer Programming## 1. Introduction

In our prior work [1], we proposed a mobility model to describe the movement pattern of MNs within a large disaster area. Moreover, we studied the coverage problem with no capacity constraints on RNs. In this paper, we assume that each RN can support a limited number of users. Then the problem to be studied is formulated as finding the deployment of a given number of RNs such that: Open image in new window most MNs can be covered; Open image in new window the network throughput can be maximized.

We first study the COverage-oriented Relay Placement (CORP) problem of deploying a set of Open image in new window RNs to cover a maximum number of MNs. As an initial setup, we consider the subproblem of Relay Assignment for COverage-oriented Relay Placement (RA-CORP) which is, for any given RNs' placement, to obtain the optimal associations between RNs and MNs using maximal matching method. Secondly, the Greedy Incremental COverage (GICO) algorithm is proposed to iteratively find the optimal location for the RN, one at each time. Thirdly, we put forward the constrained exhaustive search (CES) method to produce the optimal solution to the CORP problem as a benchmark for the GICO algorithm.

We also investigate the CApacity-oriented Relay Placement (CARP) problem to maximize the total throughput of DAWN. In this case, the Relay Assignment for CApacity-oriented Relay Placement (RA-CARP) can be formulated as the assignment problem and solved by the Hungarian method [2]. Subsequently, we propose the Greedy Incremental Capacity (GICA) algorithm to find the RNs' positions one by one. In comparison, the optimal placement of all RNs can be obtained by solving a complicated binary integer programming problem but at very high computational complexity.

The rest of this paper is organized as follows. In Section 2, related work on disaster area networks, mobility model, and base station placement in wireless networks is summarized. In Section 3, we describe the mobility model of MNs within the disaster area; Section 4 presents the problem formulation of maximizing coverage for DAWN; Section 5 presents the technical approaches to solve the CORP problem. In Section 6, we present the problem formulation of maximizing aggregate throughput for DAWN. Subsequently, Section 7 presents the technical approaches to solve the CARP problem; simulation results are given in Section 8, followed by conclusions in Section 9.

## 2. Related Work

### 2.1. Disaster Area Wireless Network

Recently, many kinds of wireless networks have been proposed to be applied to disaster area relief operations. In [3], Hiroaki et al. propose and evaluate a mobile ad hoc network system to pursue the location and personal information of victims in occurrence of disaster. In [4], Kanchanasut et al. describe an emergency communication network platform designed for collaborative simultaneous emergency response operations using a combination of mobile ad hoc networking and a satellite IP network operating with conventional terrestrial internet. In [5], Malan et al. introduce a wireless infrastructure intended for emergency medical care, integrating low-power, wireless vital sign sensors, and PC-class systems. In addition, Zussman and Segall propose to construct an ad hoc network of wireless smart badges in order to acquire information from trapped survivors [6]. Besides, a novel ballooned wireless mesh network [7] has been proposed for emergency information system. All these works either assume the majority network nodes are static or mention little about their mobility model. Therefore, they fail in constructing the disaster area communication system to accommodate dynamic node configurations.

### 2.2. Macroscopic Mobility Model

In the recent years, several different macroscopic mobility models have been proposed and used for performance evaluation of networks. The fluid mobility model [8, 9, 10] conceptualizes traffic flow of users as the flow of a fluid, which models mobility in terms of the mean number of users crossing the boundary of a given area. Derived from transportation theory, these models give an aggregated description of the movement of several users, ranging from street scale and city scale [11, 12, 13] to national and international scales [12, 14]. Furthermore, two different event-driven role-based mobility models are designed for disaster area relief applications [15, 16]. However, these two models only apply in small area with specific disaster sites.

### 2.3. Base Station Placement

There have been extensive researches dedicated to base station placement problem in wireless sensor networks. In [17], a multiobjective metric is proposed for placing multiple base stations at the optimal positions in wireless sensor networks, including coverage, fault tolerance, energy consumption, and network delay. In [18], Shi and Hou propose a (1- Open image in new window ) optimal approximation algorithm to place base station so that the network lifetime could be maximized. In [19], a polynomial time heuristic is proposed for optimal base station selection within a wireless sensor network. In [20], Pan et al. study base station placement problem to maximize network lifetime. Most of existing base station placement schemes are designed for wireless networks with nodes at specific positions. Therefore, they are not suitable for the proposed mobile scenarios.

## 3. System Modeling and Applications

### 3.1. Mobility Model

### 3.2. Network Model

We consider a set of MNs moving within the disaster area following the mobility model described previously and assume that a fixed number of RNs are ready for deploying to connect all MNs to the backbone network. We assume that all MNs have small transmission range Open image in new window . The transmission region of an MN is defined as the area in which all points are within distance Open image in new window from the MN. The Open image in new window th MN Open image in new window can communicate bi-directionally with the Open image in new window th RN Open image in new window if the distance between them Open image in new window . In other words, Open image in new window is said to be covered by Open image in new window if Open image in new window is within the transmission region of Open image in new window . RNs are able to communicate with each other without distance constraints and they form the backbone network. We assume that the relay stations can be installed on vehicles and can quickly move to any locations in the disaster area. We assume each MN occupies one orthogonal channel associated with an RN for at least one time unit to communicate bi-directionally. Since the RN has a limited bandwidth, each RN can only support a certain number of MNs. As a note, no interference issues are considered in our network model due to abundant unoccupied spectrum in the disaster area.

## 4. Maximizing Node Coverage: Problem Formulation

We first provide the notation used in this section. Let Open image in new window denote the set of four vertices of square Open image in new window . Open image in new window denotes the disk centering at Open image in new window with radius as Open image in new window . A spot Open image in new window is said to be covered by Open image in new window if Open image in new window , denoted as Open image in new window . A polygon Open image in new window is said to be covered by Open image in new window if for any point Open image in new window within Open image in new window , Open image in new window , denoted as Open image in new window .

For the CORP problem, RNs should be placed at positions to connect the maximum number of MNs. As MNs follow a macroscopic mobility model, we choose to cover the active (busy) squares instead of tracking the individual MNs. In other words, if a busy square is covered, then all MNs within the square are connected. We now claim Theorem 1 to show that a square can be covered by a circle if and only if its four vertices are within that circle.

Theorem 1 (covering a polygon).

Assume a polygon Open image in new window , where Open image in new window and Open image in new window denote the set of vertices and edges, respectively. If for all vertex Open image in new window , Open image in new window , then Open image in new window .

Proof.

First of all, we need to acknowledge the fact that if Open image in new window , Open image in new window , then Open image in new window ( Open image in new window denotes the edge connecting Open image in new window and Open image in new window ), Open image in new window (it is obviously true since the edge is fully contained in one circle if the two terminals are within the same circle). For all Open image in new window , we have Open image in new window , where Open image in new window and Open image in new window . Since for all Open image in new window , Open image in new window , then Open image in new window and Open image in new window . As a result, Open image in new window , and Open image in new window .

*feasible circle*of each square. For every busy square in the DAWN, each has a corresponding feasible circle. These feasible circles may overlap each other and the intersected areas are referred as

*shared regions*. As shown in Figure 4, three feasible circles are intersected and an RN placed in the shared region (shaded area) can cover all 3 squares at the same time. In a DAWN, these shared regions form a candidate set for RN placement. In this paper, our analysis and simulations are derived directly based on the concept of shared regions, instead of using traditional Cartesian coordinates.

The CORP problem is defined as follows. Given a set of busy squares, in which each contains some MNs inside with transmission range Open image in new window , and Open image in new window RNs each with capability Open image in new window , find the optimal placement of the RNs, such that the maximum number of MNs is covered.

## 5. Maximizing Node Coverage: Technical Approaches

In this section, we first present a maximal matching method to solve the RA-CORP problem if the RNs' positions are known. Secondly, we propose the GICO and CES algorithms to tackle the CORP problem. In addition, we conduct complexity analysis for both algorithms.

### 5.1. Relay Assignment for Fixed RN Positions (RA-CORP)

The RA-CORP problem tries to find the optimal association between MNs and RNs. We use a bipartite graph to represent the RA-CORP problem, and then use a sparse matrix-based algorithm to find the maximum-sized matching between the MNs and RNs.

Definition 1.

Denote the feasible circle of the Open image in new window th busy square Open image in new window as Open image in new window . Assume Open image in new window , Open image in new window have some area overlap, then the intersection area is defined as a *shared region* that only MNs within busy squares Open image in new window ( Open image in new window ) can access, denoted as Open image in new window , or simply as Open image in new window .

At any time, MNs are distributed within a set of busy squares. The feasible circles of these busy squares intersect and yield a set of shared regions Open image in new window where Open image in new window denotes the cardinality of the set Open image in new window . A fixed number of RNs Open image in new window are deployed at Open image in new window , Open image in new window . Each RN can support at most Open image in new window MNs within the squares covered by the RNs. Then the RA-CORP problem is formulated as

where Open image in new window denotes that Open image in new window is connected with Open image in new window and 0 otherwise. Open image in new window denotes the set of RNs that cover Open image in new window . Open image in new window denotes the set of MNs that are covered by Open image in new window . The second constraint demands that each MN can at most connect to one RN. The third constraint shows that at most Open image in new window MNs can connect to one RN.

In this paper, we use a sparse matrix-based approach [21] to find the maximal matching between MNs and channels of RNs for each RN. This approach yields the optimal solution. The complexity of finding the maximal matching within a bipartite graph is Open image in new window .

### 5.2. Relay Placement for Optimal Coverage (CORP)

We first claim Theorem 2 about complexity of the CORP problem.

Theorem 2.

The CORP problem is NP-complete.

Proof.

See Appendix .

Then we perform aggregation for all shared regions to reduce the solution space. Since the CORP problem is NP-complete, we introduce a heuristic approach GICO to solve the problem. To measure the performance of GICO, we also give the optimal solution by employing the CES algorithm.

#### 5.2.1. Aggregation

The aggregation procedure aims to reduce the cardinality of the set of shared regions, thus greatly reduces the solution space. Given a set of shared regions Open image in new window , the aggregation proceeds as in Algorithm 1. We say Open image in new window belongs to Open image in new window or Open image in new window contains Open image in new window if Open image in new window Let Open image in new window and Open image in new window denote the set of all shared regions, and reduced set of shared regions, respectively. Open image in new window denotes the cardinality of the set Open image in new window .

**Algorithm 1:** Aggregation procedure.

**Aggregation**( Open image in new window )

- 2
**for**i=2 to Open image in new window - 3
sign=0;

- 4
**for**j=1 to Open image in new window - 5
**if**Open image in new window belongs to Open image in new window - 6
remove Open image in new window from Open image in new window ;

- 7
sign=1;

- 8
**end if**; - 9
**if**Open image in new window belongs to Open image in new window - 10
sign=2;

- 11
break;

- 12
**end if**; - 13
**end for**; - 14
**if**sign==0 or sign==1 - 15
- 16
**end if**; - 17
**end for**; - 18
**return**Open image in new window ;

The procedure reduces the cardinality of the set of shared regions by removing those that are contained by others. The procedure reduces the solution space without losing any coverage options. The reason is that RNs placed in Open image in new window that contains Open image in new window can cover busy squares Open image in new window . Therefore, the set Open image in new window can contain all shared regions but with the minimal cardinality.

#### 5.2.2. Greedy Incremental Coverage (GICO)

The GICO algorithm is based on the following idea. Although it is not computationally feasible to perform an exhaustive search for placing Open image in new window RNs simultaneously, it is possible to choose an optimal position to place one RN at a time. When the RN is placed at each shared region, the optimal relay assignment can be obtained by utilizing the maximal matching method. The best shared region for placing one node can be found by exhaustively searching all shared regions in Open image in new window . Once the location for this RN is fixed, the next RN can be placed following the same procedure. It should be noted that when placing next RN, those previously placed RNs should be jointly considered for relay assignment in order to compute the coverage values. In this approach, the RNs are placed one by one until all Open image in new window RNs are deployed in the set Open image in new window .

As listed in Algorithm 2, GICO works as follows. Open image in new window represents the shared regions that have been chosen to place RNs. Open image in new window denotes the cardinality of Open image in new window . It is possible that Open image in new window contains multiple same shared regions, which means that multiple RNs should be placed in that shared region. At line 4, each shared region in the set Open image in new window is added into the current set Open image in new window and the set Open image in new window is obtained. The procedure at line 5 calculates and stores the maximal matching based on RNs' deployment according to the current Open image in new window . Open image in new window denotes the calculated optimal maximal matching while Open image in new window denotes the set of shared regions that each contains one RN. After executing the procedure of lines 4 and 5 for Open image in new window times, the best next shared region to place one RN is found (line 7) and added to the set Open image in new window (line 8). Therefore, after greedily choosing RN placement one by one for Open image in new window times, we can finally obtain the solution.

**Algorithm 2:** GICO.

**GICO**( Open image in new window )

- 1
- 2
**for**i=1 to Open image in new window - 3
**for**j=1 to Open image in new window - 4
- 5
Open image in new window

**RA-CORP**( Open image in new window ) - 6
**end for**; - 7
[maximum, index]=

**Max**(value); - 8

9**end for**;

10 **return** Open image in new window ;

#### 5.2.3. Constrained Exhaustive Search (CES)

In order to obtain the optimal solution as a benchmark for our GICO algorithm, we need to search all possible combinations of the shared regions. However, even after employing the aggregation procedure to reduce the size of solution space, the complexity for searching the optimal solution could still be as high as Open image in new window Therefore, we resort to devising the CES algorithm to further reduce the solution space by adding one constraint to the combinations of shared regions. The constraint is that for each set of Open image in new window RN placements, each RN should cover at least one MN based on the RA-CORP results. In particular, the number of RNs placed in one shared region times RNs' capability should not exceed the total number of MNs in those busy squares that are covered by the RNs by more than Open image in new window , shown as

where Open image in new window denotes the number of RNs placed at Open image in new window and Open image in new window denotes the number of MNs in the busy square Open image in new window .

### 5.3. Complexity Analysis

We first discuss the complexity of the aggregation procedure. Based on Algorithm 1, the procedure between line 5 to line 12 is iterated Open image in new window times, and the procedure between line 14 to line 16 is iterated Open image in new window times. As neither Open image in new window nor Open image in new window is influenced by Open image in new window , the complexity for the aggregation procedure is Open image in new window .

Secondly, we analyze the complexity of GICO. Based on Algorithm 2, the complexity of the GICO algorithm is Open image in new window , because the procedure from line 4 to line 5 is iterated Open image in new window times, and the worst case complexity for the maximal matching method is Open image in new window .

For CES, the complexity analysis is more complicated. According to the constraint in (2), each shared region cannot host more than a limited number of RNs. Therefore, we assume on average each shared region can host Open image in new window ( Open image in new window ) RNs, as shown in

In other words, the list of shared regions can be extended to a longer one with length Open image in new window , on which each shared region can host at most one RN. As the worst case complexity for the maximal matching method is Open image in new window , the computation complexity for CES can be stated as Open image in new window . Now we claim that the complexity of the CES algorithm is higher than GICO, as shown in Theorem 3.

Theorem 3 (complexity of GICO and CES).

The complexity of the GICO algorithm is lower than the complexity of the CES algorithm.

Proof.

Obviously, the term Open image in new window . Therefore, the complexity for GICO is less than the complexity for CES.

## 6. Maximizing Aggregate Throughput: Problem Formulation

In Section 4, we formulate the CORP problem, which is aimed to maximize the number of MNs that can be connected to the backbone network. However, the objective of the CORP problem does not address the quality of service (QoS) requirements of individual links. In other words, the deployment of RNs has to consider not only the coverage but also the QoS performance with intelligent channel allocation. Therefore, we put forward the CARP problem in the interest of enhancing the QoS performance of DAWN.

where Open image in new window denotes the area of the polygon formed by connecting the intersection points in a counterclockwise order, which can be calculated using

Based on Shannon formula, the channel capacity of a link can be expressed as (8) using path loss model

where Open image in new window denotes the capacity of the channel, Open image in new window denotes the bandwidth of the channel, Open image in new window denotes the distance between the transmitter and receiver of the link, Open image in new window denotes the path loss coefficient, Open image in new window denotes the transmit power, Open image in new window denotes the noise power, Open image in new window and Open image in new window are the transmitter and receiver antenna gains, respectively, Open image in new window denotes the wavelength of the transmitted signal, Open image in new window denotes the frequency of the transmitted signal, Open image in new window is the velocity of radio-wave propagation in free space, which is equal to the speed of light.

Since MNs follow the macroscopic mobility model, we resort to developing the two-dimensional integral (9) to compute the throughput of the link between an MN and its assigned RN. In (9), Open image in new window denotes the coordinates of the left downward vertex of the Open image in new window th busy square Open image in new window . Open image in new window denotes the throughput of the link between one MN in Open image in new window and one RN placed at Open image in new window . Open image in new window denotes the side length of each square

The CARP problem is formulated as follows. Given Open image in new window MNs each with transmission range Open image in new window that are distributed within a set of busy squares and Open image in new window RNs each with capability Open image in new window , find the optimal positions for the RNs such that the aggregated throughput of all established links between MNs and RNs are maximized.

## 7. Maximizing Aggregate Throughput: Technical Approaches

In this section, we investigate the CARP problem of deploying a set of RNs to maximize the total throughput of DAWN. We first consider the optimal relay assignment for fixed RN positions, which can be solved using the Hungarian method. On this basis, we propose the GICA approach to tackle the CARP problem. In addition, since the CARP problem falls into a binary integer programming formulation, the branch and bound algorithm [22] is adopted to produce the optimal solution as the benchmark for the GICA approach.

### 7.1. Relay Assignment for Fixed RN Positions (RA-CARP)

At any time, MNs are distributed within a set of busy squares. The feasible circles of these busy squares intersect and yield a set of shared regions Open image in new window A fixed number of RNs Open image in new window are deployed at the set of centroid points Open image in new window Open image in new window . Each RN can support at most Open image in new window MNs to access the network in the squares that it covers. Now the RA-CARP problem is formulated as

where Open image in new window denotes that Open image in new window is connected with Open image in new window and 0 otherwise; Open image in new window denotes the index of the busy square where Open image in new window is. Open image in new window denotes the index of the shared region where Open image in new window is placed. The second constraint denotes that each MN can at most connect to one RN. The third constraint shows that at most Open image in new window MNs can connect to one RN.

Given MNs placed within busy squares and RNs deployed in some shared regions, the bipartite graph can be generated as in Figure 5. It can be seen that for each MN-RN pair, there are Open image in new window edges each with a weight equal to the capacity value of the corresponding link. Note that for those pairs that the RN does not cover the MN, the edges are assigned weights equal to 0. We now can generate a gain matrix Open image in new window shown as

where Open image in new window and Open image in new window . We now present the Hungarian method [2] as follows.

Step 1.

If Open image in new window is not a square matrix, we have to augment Open image in new window into a square matrix by padding rows or columns with all zeros.

Step 2.

Multiply the matrix Open image in new window by Open image in new window .

Step 3.

Subtract the minimum value of each row from row entries.

Step 4.

Subtract the minimum value of each column from column entries.

Step 5.

Select rows and columns across which you draw lines, such that all zeros are covered and that no more lines have been drawn than necessary.

Step 6.

If the number of lines equals the number of rows, choose a combination of zero elements from the modified gain matrix such that the position of each chosen element is incident on a unique row and column. Then the optimal assignment result consists of the RN-MN pairs as represented by the chosen elements in the modified gain matrix. If the number of lines is less than the number of rows, go to Step 7.

Step 7.

Find the smallest element which is not covered by any of the lines. Then subtract it from each entry which is not covered by the lines and add it to each entry which is at the intersection of a vertical and horizontal line. Go back to Step 5.

### 7.2. Relay Placement for Maximal Aggregate Throughput (CARP)

We claim Theorem 4 about complexity of the CARP problem.

Theorem 4.

The CARP problem is NP-complete.

Proof.

See Appendix .

Since the CARP problem is NP-complete, we introduce a heuristic approach GICA to solve the problem. To measure the performance of GICA, we also present the optimal method for the CARP problem.

#### 7.2.1. Greedy Incremental Capacity (GICA)

According to Algorithm 3, the algorithm works as follows. Open image in new window denotes the set of centroid points that have been chosen to place RNs. Open image in new window denotes the set of centroid points of all shared regions. Open image in new window denotes the set of centroid points that have been chosen. Open image in new window denotes a current set of centroid points with each hosting one RN. At line 4, each centroid point in Open image in new window is added to Open image in new window and the set Open image in new window is obtained. The procedure at line 5 calculates and stores the total throughput yielded by the Hungarian method based on Open image in new window . Open image in new window denotes the calculated optimal association between MNs and RNs. After executing the procedure of lines 4 and 5 for Open image in new window times, the best next centroid point to place one RN is found (line 7) and added to Open image in new window (line 8). Therefore, after greedily choosing centroid points one by one for Open image in new window times, we can finally obtain the set of centroid points Open image in new window to place RNs.

**Algorithm 3:** GICA.

**GICA**( Open image in new window )

- 1
- 2
**for**i=1 to Open image in new window - 3
**for**j=1 to Open image in new window - 4
- 5
Open image in new window

**RA-CARP**( Open image in new window ) - 6
**end for**; - 7
[maximum, index]=

**Max**(value); - 8
- 9
**end for**; - 10
**return**Open image in new window ;

#### 7.2.2. Optimal Solution to CARP Problem

We show that the RA-CARP can be formulated as a binary integer programming problem when RNs are placed at fixed positions. Subsequently, the GICA method utilizes the Hungarian method to greedily place one RN at an iteration. Therefore, the solutions yielded by GICA cannot be guaranteed optimal because the RN assignment and placement are considered separately. It would be natural to believe that only when we search all solution space can the optimal solution be produced.

Hereby we introduce two binary variables Open image in new window and Open image in new window Open image in new window denotes that Open image in new window is connected with Open image in new window and 0 otherwise; Open image in new window denotes that Open image in new window is placed at the centroid point of the Open image in new window th shared region and 0 otherwise. Then we can jointly formulate the CARP problem as

Since the objective term Open image in new window contains the product of two variables Open image in new window and Open image in new window , it is difficult to solve. According to [23], the product of multiple binary variables Open image in new window can be substituted by a new variable Open image in new window with two constraints that ensure that Open image in new window if there exists Open image in new window , and Open image in new window if for all Open image in new window . Therefore, we transform (12) into a binary integer programming problem shown as

Now the CARP problem is formulated as a binary integer programming problem. We then utilize the branch and bound algorithm to solve it. The algorithm searches for an optimal solution by solving a series of LP-relaxation problems, in which the binary integer requirement on the variables Open image in new window is replaced by the weaker constraint Open image in new window . More details can be referred to [22].

### 7.3. Complexity Analysis

We first discuss the computation complexity of the Hungarian method to assign MNs when RNs are placed at fixed positions. According to [2], the complexity is Open image in new window .

Then we analyze the complexity of the GICA algorithm. Let Open image in new window denote the number of shared regions. Based on Algorithm 3, as the procedure on lines 4 to 5 is iterated Open image in new window times, the computation complexity of the GICA algorithm is Open image in new window .

The optimal method to the CARP problem uses the branch and bound algorithm to solve a binary integer programming problem. As the number of binary variables is Open image in new window , the worst case complexity for the optimal method is Open image in new window . It is apparent that the complexity of the GICA algorithm is much lower than the optimal algorithm.

## 8. Simulation Results

In this section, we present the numerical results obtained from the simulation using high level programming language. For the CORP problem, we compare the performance of the GICO algorithm and the CES algorithm. For the CARP problem, we compare the performance of the GICA algorithm and the optimal algorithm. It is illustrated that the two greedy algorithms both merit close-to-optimal performance and low complexity.

### 8.1. Simulation Setup

The initial configuration of CI values of squares in disaster area. The CI values are randomly chosen between the interval Open image in new window .

Square index | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
---|---|---|---|---|---|---|---|---|---|---|

1 | 2 | 10 | 10 | 5 | 9 | 2 | 5 | 10 | 8 | 10 |

2 | 7 | 1 | 9 | 10 | 7 | 8 | 8 | 4 | 7 | 2 |

3 | 8 | 1 | 3 | 1 | 1 | 9 | 7 | 4 | 10 | 1 |

4 | 5 | 4 | 8 | 8 | 2 | 5 | 5 | 7 | 8 | 8 |

5 | 3 | 7 | 7 | 2 | 2 | 5 | 10 | 4 | 6 | 3 |

6 | 8 | 3 | 6 | 7 | 9 | 10 | 6 | 2 | 2 | 3 |

7 | 9 | 3 | 9 | 3 | 10 | 4 | 2 | 3 | 7 | 5 |

8 | 4 | 9 | 6 | 6 | 10 | 3 | 8 | 8 | 4 | 6 |

9 | 1 | 1 | 6 | 8 | 10 | 2 | 6 | 5 | 1 | 4 |

10 | 2 | 8 | 4 | 6 | 2 | 7 | 3 | 7 | 7 | 8 |

### 8.2. Simulation Results for Maximal Coverage

We introduce *coverage percentage* as the measurement of coverage performance, which is defined as the ratio of the number of covered MNs to the total number of MNs working in the disaster area.

### 8.3. Simulation Results for Maximal Throughput

## 9. Conclusion

In this paper, we study the dynamic deployment of mobile relays in DAWN to enable and improve the communications for the first responders during their operations. A mobility model is used to capture the movement pattern of the MNs and their communications to the RNs. Given a fixed number of relay nodes, the optimization problem is to determine the locations of the RNs as the MNs move in the disaster area nomadically. Two performance objectives, including maximal node coverage and maximal network capacity, are considered, respectively, in this study.

In the coverage problem (CORP), the performance objective is to place the RNs that can connect the maximum number of MNs in the network. As a preliminary step, we employ a maximal matching method to find the optimal relay node assignment for the static network scenario, that is, all RNs are fixed. Subsequently, we present the greedy incremental coverage algorithm (GICO) and the optimal constrained exhaustive search (CES) algorithms. The GICO algorithm is suboptimal but with significantly less computational complexity than the CES algorithm. The simulation results show that GICO algorithm can achieve close to optimal performance at different network setup and configurations.

In the capacity problem (CARP), the performance objective is to maximize the aggregated network throughput for all MNs in the DAWN. As an initial step, we first consider the relay node assignment for the static case that can be solved using the Hungarian method. Similarly, we also present both the greedy incremental capacity algorithm (GICA) and the optimal algorithm. The optimal solution for CARP can be obtained through the binary integer programming approach but at much higher computational complexity. The simulation results show that the GICA algorithm can produce near optimal results.

In addition, it is observed that network generally yields better coverage and throughput performance for scenarios in which all MNs start from 1 corner than from 4 corners of a disaster area. The tradeoff is that it would require longer time to clear the entire disaster area. As a conclusion, we advocate using the greedy algorithms to determine the dynamic relay placement in the deployment of disaster area wireless networks, in which real-time computation is practically more important.

## Appendices

### A. Proof of Theorem 2

NP-completeness applies directly not to optimization problems, however, but to decision problems, in which the answer is simply "yes" or "no" [24]. We first present the decision problem CORP-D associated with the CORP problem as follows. Given a set of busy squares, the number of MNs in each square, the transmission range Open image in new window and Open image in new window RNs each with capability Open image in new window , is it possible to cover all MNs using exactly Open image in new window RNs? To prove the NP-completeness of the CORP problem, it suffices to prove that the decision problem CORP-D is NP-complete.

We start by arguing that CORP-D Open image in new window NP. Then we prove that the CORP-D problem is NP-hard by showing that MSC Open image in new window CORP-D, ( Open image in new window denotes a transformation of polynomial time. MSC denotes the NP-complete minimum set cover problem). Because the CORP-D problem is both NP and NP-hard, it is NP-complete.

To show CORP-D Open image in new window NP, we deploy Open image in new window RNs in the shared regions. Then to find if such deployment of Open image in new window RNs can cover all MNs is tantamount to solving the RA-CORP problem. As the RA-CORP problem has been proved to be solved in polynomial time, CORP-D Open image in new window NP.

We next prove that MSC Open image in new window CORP-D, which shows that CORP-D is NP-hard. Let Open image in new window be an instance of the Open image in new window problem, where Open image in new window denotes the collection of subsets of a set Open image in new window , Open image in new window denotes the minimum cardinality of the set Open image in new window such that Open image in new window and Open image in new window . To obtain an instance of the CORP-D problem we only need to define the capacity bound Open image in new window for each RN. Let Open image in new window be the number of all MNs. Then we build the relationship between instances of the CORP-D problem and the MSC problem as follows. Open image in new window Each element in the set Open image in new window corresponds to one MN; Open image in new window each shared region corresponds with one subset Open image in new window . Open image in new window Open image in new window is covered by the Open image in new window th shared region if Open image in new window . Then we must prove that Open image in new window RNs can cover all MNs if, and only if, there exists Open image in new window , such that Open image in new window and Open image in new window .

First, suppose that Open image in new window MNs in a set of busy squares can be covered by Open image in new window RNs, each with a capacity of Open image in new window . Then for Open image in new window deployed at one shared region, the corresponding subset is chosen as one element in Open image in new window . Since Open image in new window RNs can cover all Open image in new window MNs, it is easy to see that Open image in new window and Open image in new window .

Now assume that there exists Open image in new window such that Open image in new window and Open image in new window . For Open image in new window we can place one RN Open image in new window at the Open image in new window th shared region. In the meantime, Open image in new window covers Open image in new window if Open image in new window . Since Open image in new window all Open image in new window MNs are covered by the RNs. Since Open image in new window , the number of RNs placed is no larger than Open image in new window .

Thus we have shown that the CORP-D problem is NP-complete, which completes the proof.

### B. Proof of Theorem 4

We start by arguing that CARP-D Open image in new window NP (CARP-D is the decision problem associated with the CARP problem). Then we prove that CARP-D is NP-hard by showing that CORP-D Open image in new window CARP-D. Because the problem CARP-D is both NP and NP-hard, the problem CARP is NP-complete.

To show CARP-D Open image in new window NP, we deploy Open image in new window RNs in the shared regions. Then to find if such deployment of Open image in new window RNs can produce the objective amount of throughput is tantamount to solving the RA-CARP problem. As the RA-CARP problem has been proved to be solved in polynomial time, the problem CARP-D Open image in new window NP.

We next prove that CORP-D Open image in new window CARP-D, which shows that the problem CARP-D is NP-hard. Let Open image in new window and a positive integer Open image in new window be an instance of the CORP-D problem. To obtain an instance of the CARP-D problem we only need define the capacity of each link between one MN and RN. Let the capacity for every link between each RN and MN be 1 unit. Then we recognize the instance of the CARP-D problem the same as the instance of the CORP-D problem. This transformation surely consumes polynomial time. Subsequently, we must prove that Open image in new window units of throughput can be produced if and only if Open image in new window MNs can be covered by the same set of RNs.

First, suppose that Open image in new window MNs in a set of busy squares can be covered by Open image in new window RNs. Then for Open image in new window covered by Open image in new window , one link is established between Open image in new window and Open image in new window to produce one unit of capacity. Since Open image in new window MNs are covered, it is easy to see that Open image in new window units of throughput can be produced.

Now we assume that the overall network capacity is Open image in new window units. Then for each link between Open image in new window and Open image in new window , we render Open image in new window cover Open image in new window . As there are Open image in new window links available, each associated with one MN, Open image in new window MNs can be covered by the same set of RNs.

Thus we have shown that the problem CARP-D is NP-complete, which completes the proof.

## Notes

### Acknowledgments

This work has been supported in part by the National Science Foundation (NSF) through Award ECS-0725522 and by the Faculty Advancement in Research Awards from WPI. This work was presented in part at 3rd Intl. Conf. Wireless Algorithms, Systems and Applications, 2008.

## References

- 1.Guo W, Huang X, Liu Y: Dynamic relay deployment for disaster area wireless networks.
*Wireless Communications and Mobile Computing*2008.Google Scholar - 2.Kuhn HW: The hungarian method for the assignment problem.
*Naval Research Logistics Quarterly*1955., 2(1-2):Google Scholar - 3.Hiroaki U, Jun T, Kazuki S, Takaaki U, Teruo H: A manet based system for gathering location and personal information from victims in disasters.
*IEIC Technical Report*2005, 105(260):13-18.Google Scholar - 4.Kanchanasut K, Tunpan A, Awal MA, Das DK, Wongsaardsakul T, Tsuchimoto Y:
*A multimedia communication system for collaborative emergency response operation in disaster-affected areas.*Internet Education and Research Laboratory, Asian Insitute of Technology, Bangkok, Thailand; January 2007.Google Scholar - 5.Malan D, Fulford-Jones T, Welsh M, Moulton S: An ad hoc sensor network infrastructure for emergency medical care.
*Proceedings of the Workshop on Applications of Mobile Embedded Systems in Conjunction with the 2nd International Conference on Mobile Systems, Applications and Services (Mobisys '04), June 2004, Boston, Mass, USA*Google Scholar - 6.Zussman G, Segall A: Energy efficient routing in ad hoc disaster recovery networks.
*Ad Hoc Networks*2003, 1(4):405-421. 10.1016/S1570-8705(03)00041-6CrossRefGoogle Scholar - 7.Shibata Y, Sato Y, Ogasawara N, Chiba G: Ballooned wireless mesh network for emergency information system.
*Proceedings of the 22nd International Conference on Advanced Information Networking and Applications-Workshops (WAINA '08), March 2008, Quebec, Canada*1118-1122.Google Scholar - 8.Frost V, Melamed B: Traffic modeling for telecommunications networks.
*IEEE Communications Magazine*1994, 32(3):70-81. 10.1109/35.267444CrossRefGoogle Scholar - 9.Leung KK, Massey WA, Whitt W: Traffic models for wireless communication networks.
*IEEE Journal on Selected Areas in Communications*1994, 12(8):1353-1364. 10.1109/49.329340CrossRefGoogle Scholar - 10.Lo C, Wolff R, Bernhardt R: An estimate of network database transaction volume to support universal personal communications service.
*Proceedings of the 1st International Conference on Universal Personal Communications (ICUPC '92), September-October 1992, Dallas, Tex, USA*1-6.Google Scholar - 11.Bouchard RJ, Pyers CE: Use of gravity model for describing urban travel.
*Highway Research Record*1965, 88: 1-43.Google Scholar - 12.Lam D, Cox DC, Widom J: Teletraffic modeling for personal communications services.
*IEEE Communications Magazine*1997, 35(2):79-87. 10.1109/35.565664CrossRefGoogle Scholar - 13.Markoulidakis JG, Lyberopoulos GL, Tsirkas DF, Sykas ED: Mobility modeling in third-generation mobile telecommunications systems.
*IEEE Personal Communications*1997, 4(4):41-56. 10.1109/98.612276CrossRefGoogle Scholar - 14.Slater PB: International migration and air travel: global smoothing and estimation.
*Applied Mathematics and Computation*1993, 53(2-3):225-234. 10.1016/0096-3003(93)90103-LCrossRefGoogle Scholar - 15.Aschenbruck N, Gerhards-Padilla E, Gerharz M, Frank M, Martini P: Modelling mobility in disaster area scenarios.
*Proceedings of the 10th ACM Symposium on Modeling, Analysis, and Simulation of Wireless and Mobile Systems (MSWiM '07), October 2007, Chania, Greece*4-12.CrossRefGoogle Scholar - 16.Nelson S, Harris A, Kravets R: Eventdriven, rolebased mobility in disaster recovery networks.
*Proceedings of the 2nd Workshop on Challenged Networks (CHANTS '07), September 2007, Montréal, Canada*Google Scholar - 17.Kim S, Ko J-G, Yoon J, Lee H: Multiple-objective metric for placing multiple base stations in wireless sensor networks.
*Proceedings of the 2nd International Symposium on Wireless Pervasive Computing (ISWPC '07), Feburary 2007*627-631.Google Scholar - 18.Shi Y, Hou YT: Approximation algorithm for base station placement in wireless sensor networks.
*Proceedings of the 4th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON '07), June 2007, San Diego, Calif, USA*512-519.Google Scholar - 19.Hou YT, Shi Y: Optimal base station selection for anycast routing in wireless sensor networks.
*IEEE Transactions on Vehicular Technology*2006, 55(3):813-821. 10.1109/TVT.2006.873822MathSciNetCrossRefGoogle Scholar - 20.Pan J, Hou YT, Cai L, Shi Y, Shen S: Topology control for wireless sensor networks.
*Proceedings of the 3rd International Workshop on Discrete Algorithms and Methods for Mobile Computing and Communications (ICDCS '99), August 1999, Seattle, Wash, USA*Google Scholar - 21.Pothen A, Fan C: Computing the block triangular form of a sparse matrix.
*ACM Transactions on Mathematical Software*1990, 16(4):303-324. 10.1145/98267.98287MATHMathSciNetCrossRefGoogle Scholar - 22.Wolsey LA:
*Integer Programming*. John Wiley & Sons, New York, NY, USA; 1998.MATHGoogle Scholar - 23.Belegundu AD, Chandrupatla TR:
*Optimization Concepts and Applications in Engineering*. Prentice-Hall, Upper Saddle River, NJ, USA; 1999.MATHGoogle Scholar - 24.Cormen T, Leiserson C, Rivest R, Stein C:
*Introduction to Algorithms*. MIT Press, Cambridge, MA, USA; 2002.Google Scholar

## Copyright information

This article is published under license to BioMed Central Ltd. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.