Measure throughput of cellular and WiFi MIMO radios, part 1

As cellular and Wifi systems move from single to multiple wireless signals, handsets and the chipsets that go inside them need testing. Because of the multiple signals in a MIMO (multiple input, multiple output) system, testing handsets using a wired connection doesn’t emulate the needed test conditions.

This 3-part series will take a deep dive into the modern radio architecture and MIMO signaling. It will examine sophisticated new baseband algorithms designed to optimize throughput of 802.11n/ac and 4G systems. We will study the nature of a MIMO airlink channel and RF propagation and outline the factors that impact MIMO throughput. Finally, we will look at methods of measuring throughput and characterizing the behavior of baseband algorithms.

How MIMO radios work
Getting repeatable and consistent performance measurements of 802.11n/ac and LTE MIMO devices is, for several reasons, a monumental challenge. For example:

• Modern wireless devices are designed to adapt automatically to the changing channel conditions.
• Wireless environment constantly changes vs. time, frequency and motion of radios and reflectors.
• The time-variability of path loss, multipath, Doppler and interference often baffles the decision-making logic of the adaptation algorithms and sometimes puts radios into unintended states.

IEEE 802.11ac radios can change their data rate over three orders of magnitude, ranging from 1 Mbit/s to over 1 Gbit/s for products sold today, with a theoretical maximum rate of 6.9 Gbits/s. Data rate adaptation happens on a packet-by-packet basis in response to time-variable airlink impairments, such as signal fades or interference.

The sophisticated adaptation capability resides in the Baseband layer of the radio—the intelligent logic layer. Radios have evolved since mobile phones first emerged in the 1970s from purely analog to very sophisticated logic and this logic we will examine here.

Early radios directed the voice signal out of the microphone to modulate the RF carrier (Figure 1 left). Modern radios include a baseband layer that houses the (DSP)—a technology that changes everything about radios and enables dynamic adaptation to the time-variable channel conditions, resulting in vast improvements in throughput and range.

Figure 1. A traditional analog radio (left) would modulate a signal onto an RF carrier. In a baseband radio (right), baseband logic can make the radio adaptable to time-variable wireless channel conditions.

Today's MIMO 802.11n/ac and LTE radios have more complex adaptation algorithms than legacy SISO (Single Input Single Output) 2G/3G and 802.11a/b/g devices. While SISO devices only vary modulation, MIMO radios work with a more complex MCS (Modulation Coding Scheme).

An MCS includes the following variables: modulation, coding rate, GI (guard interval), channel width, and the number of spatial streams. Thus, MIMO adaptation algorithms have several degrees of freedom, as summarized in Table 1. A spreadsheet found here provides the formulas and computes data rates for all 802.11n and 802.11ac MCSs as they are defined in the IEEE 802.11 standard. By studying the spreadsheet, you will understand how the variables listed in Table 1 combine to set the data rate on the airlink.

Table 1. Wireless adaptation techniques typically supported by baseband logic.

The technique of transmitting multiple spatial streams in the same frequency channel is called SM (Spatial Multiplexing). SM only works under favorable channel conditions, that is, in wireless channels with low signal to noise ratio (SNR) and low MIMO correlation. MIMO modes of transmission are explained in Table 2.

Table 2: MIMO modes of transmission.

Commercial wireless chipsets typically support a subset of the MIMO modes outlined in Table 2 and some implement other proprietary modes. In the fast-changing wireless environment and with real-time decision making process, adaptation algorithms are still evolving and engineers are struggling to optimize throughput under all conditions. Any odd behavior or faults with adaptation algorithms can be difficult to catch in the field amidst constantly changing uncontrolled interference and motion of devices and reflectors.

Let’s look at the time-variable factors that impact MIMO throughput, outlined inTable 3. These are explained in detail in the octoScope white paper, Throughput Test Methods for MIMO Radios.

Table 3. Factors that impact MIMO throughput.

When considering a controlled environment wireless testbed, you can use Table 3 as a guide. In part 2, we will focus on controlled environment MIMO test beds.

Also see
Measure throughput of cellular and WiFi MIMO radios, part 2
Measure throughput of cellular and WiFi MIMO radios, part 3
Test MIMO Wi-Fi and LTE radios over the air
Verify wideband MIMO with a PXI signal conditioning solution
Making MIMO test work
Slideshow: LTE test equipment
Measurements for the new WLAN standard: IEEE 802.11ac
Use LTE channel emulation for mobile test