From: A guide for the selection of routing protocols in WBAN for healthcare applications
Sl. no. | Routing protocol | Pros | Cons | Application domain |
---|---|---|---|---|
1 | HIT [8] | Requires only 25% of the time required by PEGASIS and LEACH for data collection, network longevity is 1.05 and 1.44 times that of PEGASIS and LEACH respectively | Not mentioned for a specific medical application, the terms security, fault-tolerance and reliability of the network need to be addressed | Micro sensor networks, bio medical sensing like EEG and EMG signals, bio electric computer interfaces |
2 | AnyBody [11] | A self-organizing protocol, maintains constant number of clusters with increasing number of nodes. PDR is approximately 100% | The metrics like network delay and energy consumption are not analyzed. Hence cannot considered for critical medical applications | Periodic patient monitoring in hospitals |
3 | WASP [16] | Minimizes the coordination overhead, throughput obtained is 94%. It can minimize the delay by reducing the number of levels in the spanning tree and also the energy consumption | Mobility is not supported, hence cannot be used for dynamic sensor network applications | For indoor hospital patient monitoring |
4 | CICADA [14] | Enhanced mobility is supported, generation of the scheme is easier, end to end delay is about 110Â ms, nodes wake up only to transmit and receive data, hence dissipation of energy is minimized | It does not support traffic from the sink to the nodes | For sensors where computational resources are scarce |
5 | TICOSS [15] | Doubles the network lifetime for high traffic scenarios, PDR is higher than 92%. Lifetime of 4Â min per Joule for TICOSS with 802.15.4 and 2Â min per Joule with 802.15.4 alone, saves energy due to timezone coordinated sleeping mechanism | Not suitable for delay tolerant networks and also it is not an application specific protocol | Continuous vital sign monitoring, for ambient sensor nodes placed throughout a site |
6 | Routing service FRAMEWORK [56] | Provides prioritized routing service, user specific QoS support for small scale networks | Not considered energy consumption, which is one of the major constraints of sensor networks | Dynamic, small scale wireless body area networks |
7 | RL-QRP [55] | Uses independent distributed reinforcement learning approach for QoS route calculation, PDR above 90% | Average delay is higher (above 200Â ms), energy consumption is not considered, not sufficient for global optimization in large scale networks | Dynamic, small scale wireless body area networks |
8 | ZEQoS [37] | Suitable and effective for all data types like ordinary, delay and reliability sensitive packets, 84% consistent throughput | No considerable improvement in terms of energy consumption | Hospital BAN communication |
9 | RL-QRP [55] | Fits well in dynamic environments using optimal routing policy. Good performance during heavy traffic conditions, average delay is less than 200Â ms | Not suitable for large scale networks like multi agent systems | Dynamic bio-medical sensor networks |
10 | ENSA-BAN [25] | Along with all QoS requirements, it considers the energy consumption of nodes to improve the network performance. Approx. 96% PDR compared to DMQoS can be achieved, average delay is less than 16Â ms | Although it is a QoS aware routing protocol, the body movement is not considered | Continuous patient monitoring sensor networks |
11 | Co-LEEBA [33] | It is a link aware routing protocol. Path loss is reduced due to the use of different path loss models. With the discontinuous data transmission, it provides better life time. It maximizes the throughput to 36Â Mbps compared to other protocols with a throughput of average 2Â Mbps | Maximizes the throughput at the cost of increased delay | Implanted sensors, monitoring of aged people |
12 | DMQoS [50] | Uses modular architecture for delay critical and reliability critical packets, end to end delay is less than 120Â ms when compared to other QoS aware protocols of 260Â ms. PDR is above 92% for varying traffic flows | The estimation of several tuning parameters is not analytical. They are fixed through different simulation experiments | Resource-constrained body area networks |
13 | LOCALMOR [54] | This routing algorithm can be used along with any MAC protocol with ACK mechanism, considered the diversity of data traffic like regular, delay- sensitive, reliability-sensitive and critical traffic, end to end delay is less than 200Â ms, Packet reception ration is above 85% | Scalability of the protocol with higher number of sensor nodes should be investigated | Diverse traffic biomedical applications |
14 | TARA [59] | Handles data transmission in the presence of temperature hot spots, routes packets through low temperature area, has load balancing capability, smaller average temperature rise | Higher packet loss due to larger delay (greater than 400Â ms), unique hardware ids for nodes hence this algorithm fails to operate in id-less sensor nodes, homogeneous and not emergency supported | Implanted sensor networks and applications like retinal prosthesis and cancer detection |
15 | LTRT [58] | Optimization of routing is accomplished, very high packet delivery ratio, which is close to 100%, Smaller average temperature rise | The analysis is done only on the average temperature rise and packet loss rate | Implanted bio -medical networks, cardiac patient monitoring applications |
16 | RAIN [66] | Routes the data efficiently towards the sink in an id-less biomedical sensor networks, prevents the formation of high temperature zones in the network, maximum temperature rise increases slowly than CFLOOD protocol, PDR is greater than 90%, the average energy consumption is less than 1000 energy units compared to 3000 units of CFLOOD | Average packet delivery delay is slightly higher than CFLOOD protocol, PDR is slightly lower than CFLOOD protocol | In-vivo network of homogeneous and id-less biomedical sensor nodes |
17 | M-ATTEMPT [64] | Mobility supported, greater network lifetime (29.5%), better stability period (greater than 20%) and 29% better results for successfully received packets when compared to multihop communication, energy efficient and emergency supported | A moving node needs a new parent and the new parent may refuse this request, analysis of average/maximum temperature rise is not included | Heterogeneous and homogeneous wireless body area networks |
18 | M2E2 [61] | Mobility and multi-mode supported, energy efficient and emergency supported, throughput is above 100Mbps when compared to 50Â Mbps of M-ATTEMPT | Requires more hardware than the other protocols | Heterogeneous wireless body sensor networks |
19 | TMQoS [63] | Table-driven protocol with high network lifetime, low end to end delay which is less than 130Â ms, above 85% reliability, can meet the QoS demands along with maintaining the temperature of the nodes to an acceptable level, uses a hotspot avoidance mechanism | Average temperature rise is higher in order to meet the desired QoS demands | In-vivo wireless body area networks |
20 | ETPA [19] | Mobility supported, It solves the link disconnection problem due to body movements along with a reduction in temperature rise, PDR is up to 95% | The average delay is slightly higher than PRPLC in order to balance the temperature rise in the network | Wireless body area network with long lasting communication and scarce resources |
21 | PSR [20] | It provides reliable and secure communication against data injection attacks, PDR up to 80%, shorter routing delay | Uses ACK techniques for measuring link quality and if the number of ACKs is large, it may consume a lot of network resources, as a whole, network lifetime is less | Reliable and secure wireless body area networks |