When Designing a ZigBee Wireless Sensor Device Is It Often Necessary to Make a Choice

When designing a ZigBee wireless sensor networking device is it often necessary to make a system partitioning choice.

What makes the most sense – a ZigBee system-on-chip (SoC) that combines a 2.4GHz wireless transceiver and a processing core into a single chip or discrete approaches involving separate transceivers and host processors? 

In an SoC, the IEEE 802.15.4 compliant radio is a peripheral to the embedded processor. The packet processing and applications processing are performed within the single chip. The SoC typically includes hardware peripherals for the microprocessor to support computationally intensive functions.

In a network co-processor (NCP) design, the ZigBee stack operates on a co-processor connected to a host processor typically using a SPI or UART interface. The host is only active for those packets either sent or received by the application on that device. 

For packets that are routed, all packet processing including security processing is done on the network processor without interrupting the host processor. The impact of SPI or UART processing time would only be expected at the source or destination of a packet.

System Partitioning

A ZigBee transceiver includes only the RF transceiver and the timing-critical MAC/PHY functions, while a host processor supports the upper layers of the MAC, network protocol and applications code. 

All packets must be moved to the host for processing. Those packets that are only to be routed are moved to the host and back to the radio for retransmission typically though a UART or SPI port. AES encryption operations often are included on the transceiver; therefore, additional UART or SPI transfers are required for security processing. 

Consider transceiver performance in terms of throughput and latency. Throughput is a measure of how much data traffic the network can support and is a key metric because it determines network scalability. 

Latency is a measure of how quickly messages are passed between nodes and is critical because it determines network responsiveness. 

Throughput and latency are both functions of device partitioning and must be considered in the system architecture.

ZigBee is a hybrid mesh networking protocol that includes a backbone of routers that are always active and end devices that are normally asleep. The routers are responsible for passing messages between the end devices or from end devices to a central controller.  

Network throughput and latency are functions of how quickly routers can process data packets and forward them to the proper destination.

Router efficiency is a function of system partitioning. For SoC or NCP-based systems, routing is handled without waking or interrupting the host. Packets are typically forwarded within 5-10ms. For transceiver-based systems, the transceiver must wake or interrupt the host to process each packet. 

The wake or interrupt latency can be greater than 100µs. Packet data also must be transferred between the transceiver and host processor. 

ZigBee packets can be up to 127 bytes so transferring a packet to the host and back to the transceiver can take 0.5-4ms at typical SPI/UART data rates. 

For example, for a 5-byte payload where AES encryption is supported in the ZigBee transceiver, the latency of a single hop is 10ms in a network using an SoC and 20mse using a separate transceiver. 

Since it takes each node twice as long to process a packet, the throughput of the network using transceivers is reduced by 50%, which reduces the maximum number of devices that can be supported by 50%. 

For timing-sensitive applications such as lighting, increased latency limits the maximum number of hops allowed, reducing network scalability and reliability.

Power Consumption

The ZigBee protocol was designed to allow sleeping end devices to be in control of their battery life. Sleeping end devices set their own schedule for waking and interacting with the network, allowing designers to determine the proper balance between battery life and data updates. 

The protocol does not require resynchronisation of the sleeping end device when it wakes so the data transfer to its parent is very efficient. 

The main power consumption metric is end-node battery life. A battery-powered end node is typically asleep and wakes periodically to check for network data. When the end node is asleep, current consumption is dominated by leakage current.

If the network has data for the end node, the time required to send the data from the router to the end node depends on system partitioning. If the router is an SoC or NCP, the data request can be processed internally, and the router typically responds within 2-3 ms. 

If the router uses a transceiver, the transceiver must wake or interrupt the host processor, wait for it to create the data packet and receive the packet over the serial port, which can add up to 10 ms of delay. During this delay, the end node receiver must remain active, which can significantly reduce battery life. 

For end devices and routers that do not require a host processor, the single-chip wireless SoC system partitioning approach delivers the best network performance, the lowest power and the lowest overall cost. 

If a host processor is required by the system, the NCP system partitioning approach delivers the best performance and lowest power with the least impact on the host processor performance.