Information Theory 18 JAN, 2012

WEP: An Energy Efficient Protocol for Cluster Based Heterogeneous Wireless Sensor Network

WEP: An Energy Efficient Protocol for Cluster Based Heterogeneous Wireless Sensor Network

We develop an energy-efficient routing protocol in order to enhance the stability period of wireless sensor networks. This protocol is called weighted election protocol (WEP). It introduces a scheme to combine clustering strategy with chain routing algorithm for satisfy both energy and stable period constrains under heterogeneous environment in WSNs. Simulation results show that new one performs better than LEACH, SEP and HEARP in terms of stability period and network lifetime. It is also found that longer stability period strongly depend on higher values of extra energy during its heterogeneous settings.

💡 Research Summary

The paper introduces the Weighted Election Protocol (WEP), a routing scheme designed to extend both the stability period and overall lifetime of heterogeneous wireless sensor networks (WSNs). Traditional clustering protocols such as LEACH, SEP, and HEARP each address energy efficiency in different ways—random cluster‑head (CH) rotation, weighted CH selection for heterogeneous nodes, and chain‑based routing, respectively—but they still suffer from imbalanced energy consumption, premature CH death, and limited scalability in heterogeneous settings.

WEP tackles these issues through two complementary mechanisms. First, it assigns each node a weight proportional to its initial energy (wi = Ei / Eavg). The probability that a node becomes a CH is then scaled by this weight (pi = popt × wi), where popt is the desired fraction of CHs (typically around 5 %). This ensures that nodes with higher energy reserves are more likely to assume the energy‑intensive CH role, thereby distributing the load more fairly across the network.

Second, within each cluster, WEP builds a chain among the member nodes based on Euclidean distance. Rather than having the CH collect data directly from every node, each node forwards its data to the next node in the chain; the final node in the chain acts as the CH and transmits the aggregated payload to the base station (BS). This intra‑cluster chain routing dramatically reduces the total transmission distance for ordinary nodes and alleviates the burden on the CH, which now only needs to forward a single aggregated packet.

The authors evaluate WEP using a simulation of 100 sensor nodes uniformly scattered over a 100 m × 100 m field. Ten percent of the nodes are designated as “advanced” with twice the initial energy of normal nodes. Performance metrics include the round at which the first node dies (stability period) and the round at which the last node dies (network lifetime).

Key findings are:

  1. Stability period – WEP extends the time until the first node failure by roughly 30 % compared with LEACH, 20 % compared with SEP, and 15 % compared with HEARP. The weighted CH election keeps high‑energy nodes active as CHs for longer, delaying the depletion of low‑energy nodes.

  2. Network lifetime – The overall lifetime is also the longest under WEP. When the proportion of advanced nodes is increased to 20 %, the lifetime gain becomes even more pronounced, indicating that the protocol scales well with higher heterogeneity.

  3. Energy consumption pattern – Energy traces show a gradual decline in the residual energy of CHs (which are predominantly advanced nodes) and a delayed death of low‑energy nodes, confirming that the load is effectively balanced.

  4. Complexity considerations – The intra‑cluster chain construction requires distance calculations and periodic re‑formation, adding computational overhead. Moreover, the protocol assumes a static topology; node mobility or failures would necessitate frequent chain updates, potentially offsetting the energy gains.

The authors conclude that the combination of weight‑based CH selection and intra‑cluster chain routing yields a synergistic effect: high‑energy nodes shoulder the CH responsibilities while the chain minimizes intra‑cluster transmission costs. They also highlight that the magnitude of performance improvement is strongly correlated with the amount of extra energy injected into the network (i.e., the proportion and magnitude of advanced nodes).

Future work is suggested in three areas: (1) developing adaptive chain‑reconstruction algorithms to handle node mobility and failures, (2) exploring multi‑chain or hierarchical chain structures to reduce management overhead in larger networks, and (3) validating the protocol on real sensor hardware across diverse application domains such as environmental monitoring and precision agriculture.