This work considers the problem of minimizing the amount of communication needed to send readings from a set of sensors to a single destination in energy constrained wireless networks. Substantial gains can be obtained using packet aggregation techniques while routing. The routing algorithm we developed, called data funneling, allows the network to considerably reduce the amount of energy spent on communication setup and control, an important concern in low data-rate communication. This is achieved by sending only one data stream from a group of sensors to the destination, instead of having an individual data stream from each sensor to the destination. This strategy also decreases the probability of packet collisions when transmitting on a wireless medium because the total number of packets is reduced by incorporating the information of many small packets into a few large ones. Additional gains can be realized by efficient compression of data. This is achieved by losslessly compressing the data by encoding information in the ordering of the sensors’ packets. This “coding by ordering” scheme compresses data by suppressing certain readings and encoding their values in the ordering of the remaining packets. Using these techniques together can more than halve the energy spent in communication.
In the quest to implement a network of low-power, reconfigurable sensor nodes, it is necessary to aggressively scale the amount of energy each block requires while maintaining full functionality, network connectivity, and data throughput. The RF transceiver block is crucial, as the power dissipation of this block could easily eclipse the entire sensor node power budget if not properly designed. This research focuses on the implementation of an energy efficient RF transceiver for PicoRadio.
There are three key requirements of the RF transceiver. To facilitate low-power communication of bursty sensor node data, the chosen performance metric of the transceiver is the energy required to transmit each bit of data (energy/bit). Second, in order to achieve a low-cost and small form-factor sensor node, high integration of the RF transceiver is important. Finally, because indoor sensor-node environments typically present narrowband fading (varying degrees of attenuation of narrow frequency bands), some mechanism of fading immunity is required.
To meet these requirements, a break from the traditional low-power, narrowband radio design paradigm is necessary. Some key enablers are recent developments in microelectromechanical (MEMs) technology. By utilizing MEMs resonators, it is possible to perform passive frequency translation and filtering. Traditionally, these operations are accomplished with active circuitry (i.e., mixers), consuming large amounts of power in the process. Secondly, the ultimate goal of this MEMs technology is the full integration with active circuitry. Additionally, they allow detection of widely spaced frequency bands, facilitating the design of a fading resistant architecture.
The ultimate goal of this research is a fully integrated, ultra low-power RF transceiver suitable for wireless sensor node applications.
The topology of a randomly deployed sensor network can have a dramatic impact on the network's performance, as well as its lifetime. In a radio network without topology control, every node uses the same transmission power to send packets to its neighbors, with a neighbor defined as any node that can be directly reached by this node.
This approach has some major problems. In many envisioned real world scenarios, the positions of the sensors in a network can't be pre-arranged. In those areas where node density is high, one node may have many neighbors, while in other areas where node density is low, node degree is very small. Nodes with fewer neighbors may become bottleneck points and be biased by the upper layer routing protocol to relay more traffic, die fast, and disconnect the network quickly. Those high-degree nodes will spill out unnecessary energy and generate a lot of interference to their neighborhoods.
In this research we propose a novel distributed topology control protocol called zone-based topology control. It is a two-phase procedure to generate a topology where every single node in a network has roughly the same number of neighbors. We can demonstrate a substantial energy savings from these results.
At this time we are evaluating the performance and stability of the proposed protocol in real world senarios.
Recent advancements in wireless communications and electronics have led to the development of sensor networks. Applications for sensor networks can be found in many different areas, including military, office, and home. In sensor networks, unlike in ad hoc networks, the most critical factors are not bandwidth efficiency, packet throughput, or latency, but power efficiency and scalibility. These different emphases make the design choices over the protocol stack in sensor networks very different from in ad hoc networks.
In this project, we build a power model including both the physical layer and datalink layer, with various wireless channel models assumed. We consider power consumption in actual circuit components as well as from a communication prospective (SNR/BER). Taking circuit complexity into account when considering power consumption, we found that the traditional way of designing communication systems (modulation/demodulation, MAC protocol, error control coding, etc.) does not necessarily lead to the optimum result. In this project, we intend to analyze the tradeoffs between low power and acceptable QoS in a sensor network, integrate across both the physical and datalink layer, and find the optimum operating point.
The PicoRadio Project at the Berkeley Wireless Research Center (BWRC) is focusing on networks of extremely small self-powered nodes communicating via radio. A PicoRadio network must be self-configuring, power-aware, and scalable, so the media access and network protocols are necessarily complex. The goal of the project is to build a single-chip version of a node in which much of the protocol implementation is to be hard-coded on the IC.
In order to ensure that the protocols work before the chip is built, we have implemented the protocol suite on the PicoRadio test bed: a network of nodes built from off-the-shelf components with configurable elements such as a processor and programmable logic. The test bed emulates fairly closely the conditions expected when the single-chip nodes are deployed, so we can gain a detailed understanding of the functionality of the protocols, the behavior of the network under dynamic conditions, and the parts of the protocol suite that should be parameterized.
The test bed implementation has already proven to be a valuable tool for understanding network behavior. The implementation team has performed a large set of measurements under varying conditions, and is continuing to gather data. Lessons learned from previous studies have been incorporated into the protocol set, for instance, the stability of the media access protocol in the presence of rapidly changing radio channels has been greatly improved. These lessons are currently being applied to the implementation of a preliminary version of the final PicoRadio node.
4Visiting Researcher, Swiss Federal Institute of Technology, Lausanne
Scaling of CMOS technology poses significant difficulties in precise process control and circuit operation noise reduction. In order to continue the silicon success in the nanometer regime, it is critical to explore design solutions to handle the performance variability at the early stage. Our work aims at building a cohesive process and design co-optimization framework for future technology generations. By developing a set of predictive technology and circuit performance models, current efforts are focused on investigating the impact of variations on different digital circuit designs at both gate level and micro-architecture level.
This work considers the problem of localization in low-cost, wireless sensor networks. Localization refers to the process by which the nodes in a sensor network discover their geographical location. In a typical sensor network only a few nodes know their position a priori. Thus, I am researching localization algorithms by which the unknown nodes in the network can calculate their positions using information from the known nodes. A good localization solution has the following requirements. First, it must be scalable and distributed. Second, a solution has to be tolerant to errors in the distance measurement. (Errors will be present in the distance measurements due to the difficulty of accurately measuring distance in environments with multi-path reflections.) Third, the solution must converge to accurate location estimates in many different and time-varying environments with constrained communication and computation resources. In other words, a robust solution is required.
The overall goal of this project is to research and evaluate several existing localization algorithms, then extend these algorithms to increase their robustness and incorporate features such as current error estimation, early termination when error bounds are low, and better convergence guarantees.
PicoRadio is a ubiquitous network of ultra low-power and low-cost wireless sensor nodes. Each node contains a complete protocol stack based upon the OSI model. My research is on implementing the medium access control (MAC) layer of this stack. Its functionality includes initialization and maintenance management, such as assigning local address, keeping one-hop network topology, and processing control broadcast messages and datapath between the network layer and physical layer for broadcast and unicast messages with FIFO buffers. MAC also uses CSMA and beaconing mechanisms. It discovers neighbors in the initialization phase, maintains a neighbor lookup table, forwards the neighbor information to the network layer, keeps a list of two-hop neighbors’ addresses, assigns its own local address randomly, computes its own location, maintains on a periodical basis, issues random backoff, designates channel for transmission, etc. In order to capture the functionality of the MAC, Stateflow in Simulink is chosen as the implementation tool. The underlying model of computation is suitable for simulation of control-dominated applications like the MAC. We have been focusing on functional simulations for a PicoNode network in Simulink for the purpose of protocol validation and verification. Current effort has been done on system design using Simulink Stateflow. Issues in the development of the MAC layer such as power management, power control, and CSMA sensing method are considered. Stateflow has been converted into VHDL using SSHAFT Translator. VHDL verification will be done in BEE. Automatic synthesis using SSHAFT design flow down to ASIC will be used to complete this project.
Figure 1: Design flow
My research introduces a framework where models for components in the data link layer are brought together with models for the network layer and the channel. This framework includes all the factors that influence the design of the data link layer and enables us to study the design of any component in the data link layer in the context of other components and layers. As a case study, we have built models for commonly used MAC/Link designs. We applied Banach’s fixed point theorem to solve the closed loop problem we encountered. We verified our models with network simulations using OMNET++. We also studied the impact of some important parameters.
This research addresses the algorithms and implementations for digital baseband timing recovery in wireless receivers. Timing recovery refers to the estimation and tracking of several non-idealities in the received signal caused by (1) the wireless channel itself, and (2) the RF and analog circuits in the transmitter and receiver. Parameters to be estimated include: (1) frequency, (2) phase, (3) sampling instant, and (4) gain, including multipath and scattering effects. This research looks specifically at timing recovery performed on the baseband signal (after down-conversion from the carrier) in the digital domain (after the analog to digital converter) and is particularly concerned with lowering the power consumption of the total receiver.
Digital baseband timing recovery can ease the design of the analog and RF circuitry by correcting for non-idealities caused by sub-optimal implementations. This tradeoff becomes especially important in single-chip radios when the RF and analog circuitry needs to be implemented in an ostensibly digital process with low voltages--a difficult task. By transferring some of the complexity to the digital domain, it is conjectured that the entire system can consume less power. This work is taking place within the PicoRadio project where low power is the primary goal. We investigate the architectural and implementation issues related to building low power baseband timing recovery systems in VLSI.
In this research, the computational hardware requirements for timing recovery on the various PicoRadio physical layers provide a platform for evaluation of the digital baseband timing recovery systems. The past accomplishments and ongoing efforts include modification of algorithms, and the efficient mapping of these algorithms into architectures and VLSI implementations that provide the final measure of complexity and power consumption.
This research supports the PicoRadio project at the Berkeley Wireless Research Center. This project is focused on developing an extremely low-power, wireless sensor node capable of collecting data from the environment and transmitting it over an ad-hoc multihop network.
Phase III of this project involves system level architecture options to meet the aggressive 100 µW average power consumption requirement. Work to this end falls into two categories: power budgeting and architectural choices. A preliminary power budget for the PicoNode III has been created through analysis of the expected data rates, clock frequencies, projected 0.12 µ process characteristics, and discussions with the PicoNode subgroups. As more refined microarchitectures are explored for the various subcomponents, this power budget will be revised to provide a more accurate account of the power consumption.
The second category involves architectural options for low-power operation of the PicoNode. Known techniques for low-power design include clock gating, dynamic frequency scaling, multiple supply voltages, and dynamic threshold voltage scaling. Coupled with the event driven nature of sensor and communication networks, blocks can be powered down completely while waiting for events. Clearly the power management can no longer be an afterthought of the design, since it affects the partitioning of the system, design of the individual components, and interactions between them. One current vision uses a distributed power management scheme that would activate blocks upon the receipt of an event. However, while this strategy may minimize the active power of a particular block, it may not yield the globally minimal power consumption for the entire system due to the overhead of entering and exiting the power-down modes. In situations where data moves predictably through the system, a centralized controller can be used to optimize the power characteristics of these specific scenarios. Another current research topic involves maintaining the state of the system while using aggressive power reduction schemes in sleep mode.
The goal of the PicoRadio project is to build a wireless sensor network that is versatile, self-organizing, dynamically reconfigurable, and multi-functional. The primary constraint in building the nodes is the extremely low energy budget. This imposes tight constraints on the entire design, and instead of the traditional layered protocol design, we are tightly integrating the protocol stack vertically for maximum energy optimization. The network layer of the PicoNode protocol stack has two primary functions: routing and addressing of nodes.
Addressing of nodes is based on the geographical positions of nodes. Queries in sensor networks are typically not concerned with a particular node, rather they are directed to any nodes which satisfy particular criteria of position, type, etc. This leads to our concept of class-based addressing, which is a triplet consisting of location, node type, and node sub-type. This eliminates the overhead of assigning and maintaining fixed addresses across the network as well as the problem of distributing them dynamically.
For the routing of packets across the network, we have developed an energy efficient protocol we call "energy aware routing." This scheme takes the view that trying to optimize every route to consume the least amount of energy is not in the best interest of the network. Instead, we try to optimize the power of the entire network and in the process maximize the lifetime. Thus, we find a set of routes on demand and choose between them in a probabilistic fashion. The choice is based on a metric which combines both transmit/receive power and residual energy at a node. This means that all nodes across the network participate more actively in routing, and a few nodes do not die out just because they are in a good position to forward packets. Simulations verify this behavior and the network lifetime also increases substantially.
In addition, we are working on defining the capacity of sensor networks at the network level. This capacity is basically determined by a set of nodes that are the most energy constrained, forming a cutset. The challenge is to find a routing scheme that can route packets in such a way as to minimize the energy consumption of nodes in this set. The probabilistic scheme discussed earlier performs this task to some extent, but more work is needed to actually find the optimal scheme.
In order to achieve the low power goals of PicoRadio, new architectures for the RF receiver must be researched. An important component of the receiver is the low noise amplifier. For our application, the low noise amplifier must provide high gain and adequate noise and linearity while consuming minimal power.
In this research, a design utilizing an inductively degenerated common source amplifier utilizing a RF MEMS FBAR resonator was explored. The FBAR resonator is capable of providing a high Q tank and narrowband filtering. Another advantage of the resonator is that it can ultimately be integrated on-chip. In this architecture, the resonator will be used for tuning the output tank, as well as providing high impedance at resonance in order to generate gain. On-chip spiral inductors are used at the source for input impedance matching, and in parallel with the FBAR resonator at the output to provide DC bias current through the transistors. The gate inductor used to determine the resonant frequency is implemented off-chip. Simulations have shown that a voltage gain exceeding 30 dB can be achieved while using only 500 mA from a 1.2 V supply.
To characterize the performance of the LNA, a prototype was fabricated in a 0.13 mm CMOS process. A PCB board was also made to test the LNA, and results will be available soon.
We aim to develop an efficient OS for complex real time, power-critical, reactive systems implemented on advanced heterogeneous architectures. Event-driven OS, developed specifically to target reactive event-driven systems, is much more efficient than traditional general-purpose OS. TinyOS, an existing event-driven OS, offers some very attractive concepts, but is insufficient to fulfill the ambitious software management role demanded. To overcome the limitations of TinyOS, we proposed an event-driven hierarchical power management framework. The hierarchical structure enhances design scalability, supports concurrency in both the application domain and architecture and enables power control at various granularities. The software management framework implements a hybrid power control policy and has the ability to formally devise and implement an optimal power scheduling policy.
Low power locationing systems are essential parts of distributed sensor networks. As a part of PicoNode 3 digital protocol processing chip, a locationing block is being implemented on silicon. Hop-counts from certain sensor nodes (coined anchors) with known positions are utilized to estimate the position of the node. The block executes the LS position estimation, also called triangulation, as well as encoding and decoding of the Pico Radio packets that contain locationing information. In future work the actual distances, instead of hop counts, between nodes are to be measured using radio signals. This scheme is planned to utilize the time of flight measurements of the radio signals with an accurate version of GPS-type signaling.
As a part of the PicoRadio project to build a ubiquitous ad-hoc wireless sensor node network, the physical layer has to provide a reliable point-to-point radio link under very tight power constraints. The analog transceiver building blocks make up a large percentage of the overall power budget. In order to minimize the amount of energy needed to convey one bit of information, new strategies have to be employed which take into account the power consumption not only in a communication theoretic sense in the form of energy transmitted over the channel, but also the energy needed to meet performance requirements of the analog and digital building blocks in the receiver chain. Following this approach, the PicoRadio RF group is implementing a transceiver utilizing the least number of analog components possible together with promising new technologies like RF-MEMS . To evaluate the performance, my research focuses on modeling the radio link, including these blocks. I developed a behavioral simulation framework for a point-to-point link and further abstracted the design into an analytical model. From this model, we can derive performance specifications for the blocks under development and can use them in the design process.