If you wish to contribute or participate in the discussions about articles you are invited to contact the Editor

Receiver Operations: Difference between revisions

From Navipedia
Jump to navigation Jump to search
No edit summary
m (Included editor logo.)
 
(24 intermediate revisions by 5 users not shown)
Line 1: Line 1:
{{Article Infobox2
{{Article Infobox2
|Category=Receivers
|Category=Receivers
|Title={{PAGENAME}}
|Editors=GMV
|Authors=GMV
|Level=Intermediate
|Level=Medium
|YearOfPublication=2011
|YearOfPublication=2011
|Logo=GMV
|Logo=GMV
|Title={{PAGENAME}}
}}
}}
In the first years of GNSS, when only GPS was available for public use, a receiver was not expected to perform as the navigation devices we see today. In fact, a first-generation GPS receiver could be designed to only process 4 or 5 signals at any given time, and it was deemed suitable for positioning applications. Today, with the increased availability and potential of different GNSS signals and constellations, any receiver is expected to support at least 10 or 12 channels up to hundreds, resulting in a solution with improved performance. However, at their core - and despite the many differences in target applications or implementation philosophy - most receivers still share common control and processing flow management properties, resulting in common operations<ref><i>Borre, K. et al, "A Software-Defined GPS and Galileo Receiver - A Single-Frequency Approach", Birkhäuser, chapter 5</i>.</ref> performed. The following sections describe these common operation modes in which a receiver functions. 
==Overview==
One thing most GNSS receivers have in common is how they operate in terms of processing chain, from reception of signals to solution outputs. Although their [[Receiver Types|types]], [[System Design Details|architectures]], and applications may vary, the operations (illustrated in Figure 1) apply transversely.


In the first years of GNSS, when only GPS was available for public use, a receiver was not expected to perform as the navigation devices we see today. In fact, a first-generation GPS receiver could be designed to only process 4 or 5 signals at any given time, and it was deemed suitable for positioning applications. Today, with the increased availability and potencial of different GNSS signals and constellations, any receiver is expected to track at least 10 or 12 signals in parallel channels, and up to hundreds, resulting in a more accurate solution. However, at their core - and despite the many differences in target applications or implementation philosophy - most receivers still share common control and processing flow management properties, resulting in common operations performed. The following sections describe these common operation modes in whitch a receiver functions.
[[File:Receiver_operations.png|center|thumb|650px|'''''Figure 1:''''' GNSS receiver operations diagram.]]


==Background==
After start-up procedures (e.g. loading necessary data, initializations and resource allocations), the antenna and the front-end blocks start providing digitized data continuously to the signal processing channels, also referred to as [[Baseband Processing|baseband processing]].
A GNSS receiver main objective is to determine its distance to a set of visible satellites, by tracking their transmitted signals. These signals have and underlying periodic code modulation that is precisely time-tagged by each satellite. Since the satellite time reference is [[An intuitive approach to the GNSS positioning|very accurate and the same]] for all satellites, a receiver can determine the time it took for each signal to arrive, thus determining its relative position to each satellite. This process translates into a <b>pseudorange</b> measurement, which is a rough estimate of the user-satellite distance (not compensated for signal transmission delays and clock offsets).


Besides the pseudorange, other measurements are extracted in successful tracking operations: the <b>carrier phase</b>, defined as the phase difference between the received and the internally generated carrier (codeless), is a measurement of the cycle turns (accumulated wavelengths) tracked by the receiver. This observation is very precise, but ambiguous, since the total number of integer wave cycles from transmission to reception is unknown. Also, the <b>Doppler frequency</b> seen by the receiver, caused by the dynamics of the satellite-user system, represents the difference between received and internally generated signals, and is a good representation of the relative velocity between satellite and receiver.
==Channel Level==
A typical GNSS receiver controls independent channels that are assigned to a specific signal from a specific satellite. Each channel can be in two modes: acquisition or tracking. In acquisition mode, each channel conducts a rough [[Generic Receiver Description|2D search]] (code delay and Doppler frequency) of the signal in order to assess whether the signal is present or not. The channel remains in acquisition mode until a signal is detected and a first estimate of the code delay and Doppler frequency is computed. At that point, the channel goes into tracking mode where the estimates of the code delay and the Doppler frequency are continuously refined and the navigation message is decoded.


As a result, GNSS receivers operate in a way that these observables are continuosly determined and refined in order to track the satellites, demodulate the navigation message and data, and compute the desired solution as accurately as possible.
In tracking mode, the channel monitors the quality of the tracking results in order to assess if the signal is present and is being correctly tracked. If the [[Lock Detectors|lock detection]] indicators fall below a threshold, the signal is considered lost and the channel goes back to acquisition mode and re-starts the whole process. The signal can be lost due to several factors such as shadowing of the satellite, cycle slips on the signal, or simply due to noisy initial estimates provided by the acquisition mode.


==Operations==
==Receiver Level==
One thing most GNSS receivers have in common is how they operate in terms of processing chain, from reception of signals to solution outputs. Alhough their [[Receiver Types|types]], [[System Design Details|architectures]], and [[Applications Processing|applications]] may vary, the operations (ilustrated in Figure 1 from top left to bottom right) apply transversly:
At receiver level, the outputs of the channels (measurements and navigation message) are processed into usable solutions. Depending on the application, a GNSS receiver may need measurements from a minimum of three satellites (e.g. for 2D positioning), four satellites (e.g. for 3D positioning) or more satellites (e.g. for [[Criticality of GNSS Applications|liability critical applications]]).
Therefore at global level, if the receiver has enough measurements to provide application data it outputs a solution, otherwise it remains in idle mode. Some receivers might cope with instantaneous lack of measurements by extrapolating from previous measurements or even [[Receiver Types|using information from external sources]] to support this extrapolation. An example is the possibility to use measurements from an Inertial Measurement Unit (IMU) in order to continue estimating positioning, even in degraded mode.
Furthermore, the receiver may also use measurements from a satellite with bad quality indicators just to provide a solution, even if degraded, e.g. in case availability is more important than accuracy for its specific application. This is often the case for receivers that provide 2D solutions when only measurements from three satellites are available, e.g. in [[:Category:Autonomous Applications|car applications]] where it is reasonable to assume that the user positioning can be map matched to a given road and therefore availability is the key performance factor.


[[File:Receiver_operations.png|right|thumb|650px|'''''Figure 1:''''' Typical GNSS receiver operations diagram.]]
==Advanced Operations==
Once the receiver is turned on, there is a <b>start-up</b> sequence needed to ensure that each channel is managed depending on the current operation status, and resources are allocated as needed. Before a receiver starts processing the samples from the RF stage, this initial operation mode sets up the necessary data and processes to ensure the best performance possible, also depending on the available <i>a priori</i> information (e.g. previous [[GNSS_signal|almanac or ephemeris]] information).
On top of the basic operations at channel and receiver level, the overall receiver can further benefit from information either collected from past operations or recently decoded. For example, the receiver can use the information conveyed in the navigation message of a given satellite – e.g. [[GNSS signal|almanacs]] from other satellites – in order to estimate which satellites are visible and therefore assign channels to those satellites that have a higher probability to be in view.
Following these principles, acquisition is often sub-categorized in: cold, warm and hot start mode.


As start-up and configuration is concluded, the antenna and the front end blocks start providing digitized data continuously to the signal processing channels. This input data can be in the form of real IF samples, or baseband (real and complex representation of the signal in [[Baseband Processing|I and Q components]]), and the receiver starts operating in signal processing for acquisition and tracking, with <b>no PVT solution</b> available.
*<b>In cold start</b>, the receiver has no prior information about its own position or which satellites are in view. As a consequence, each channel has to perform extensive search of all possible signals / satellites over all possible code delay / Doppler frequency pairs. It is only when (at least) one full navigation message is decoded that the receiver can infer the remaining satellites in view (through almanac information), and may redirect the other channels to follow them.


Each receiver channel acquires and tracks an input signal, by means of several [[Correlators|correlations]] between the input signal and a local replica of the carrier modulated with the PRN code. These correlations, together with the [[Tracking Loops|tracking loops]], provide sufficient information on the the satellite's motion relative to the user, and code delay and carrier phase information is provided, yielding successful synchronisation with the navigation message. The extraction of navigation data begins, and a <b>degraded PVT solution</b> can be provided as enough data is available: for instance, if only 3 satellites are tracked, a 2-D solution can be assessed, assuming a contant known height.
*<b>In warm start</b>, the receiver has access to rough initial position (e.g. last position before powering down the receiver), and to almanacs relatively up to date (e.g. via external sources or stored at power down). With this information, the receiver is able to predict which satellites are most likely to be in view, and estimate their rough code delay and Doppler frequency, hence being able to narrow down the acquisition search space.


As the navigation messages are decoded from each satellite signal, almanac and ephemeris data is continuously gathered in the receiver, providing more and more information on the behaviour of the GNSS constellation as a whole (at least until the full almanac is decoded). This information can be used to inform the acquisition process of which satellites are theoreticaly in view, enabling the search space to be refined. Also, from the navigation message, ephemeris for each satellite are continuously updated, and receivers can monitor and refine several correction computations to apply to the pseudoranges. When this process is successfull for a large enough period of time, and for enough satellites, the extracted information is used to periodically compute the <b>optimal PVT solution</b>.
*<b>In hot start</b>, the receiver has access not only to its rough initial position, but also to the ephemerides of the satellites, hence greatly reducing the acquisition search space, improving the time to acquire and therefore the final Time To First Fix (TTFF).


==Related articles==
==Related articles==
For further details on GNSS receivers, please visit the following links:
*[[Generic Receiver Description]]
*[[Generic Receiver Description]]
*[[Receiver Types]]
*[[Receiver Types]]
*[[Applications Processing]]


==References==
==References==

Latest revision as of 16:34, 18 September 2014


ReceiversReceivers
Title Receiver Operations
Edited by GMV
Level Intermediate
Year of Publication 2011
Logo GMV.png

In the first years of GNSS, when only GPS was available for public use, a receiver was not expected to perform as the navigation devices we see today. In fact, a first-generation GPS receiver could be designed to only process 4 or 5 signals at any given time, and it was deemed suitable for positioning applications. Today, with the increased availability and potential of different GNSS signals and constellations, any receiver is expected to support at least 10 or 12 channels up to hundreds, resulting in a solution with improved performance. However, at their core - and despite the many differences in target applications or implementation philosophy - most receivers still share common control and processing flow management properties, resulting in common operations[1] performed. The following sections describe these common operation modes in which a receiver functions.

Overview

One thing most GNSS receivers have in common is how they operate in terms of processing chain, from reception of signals to solution outputs. Although their types, architectures, and applications may vary, the operations (illustrated in Figure 1) apply transversely.

Figure 1: GNSS receiver operations diagram.

After start-up procedures (e.g. loading necessary data, initializations and resource allocations), the antenna and the front-end blocks start providing digitized data continuously to the signal processing channels, also referred to as baseband processing.

Channel Level

A typical GNSS receiver controls independent channels that are assigned to a specific signal from a specific satellite. Each channel can be in two modes: acquisition or tracking. In acquisition mode, each channel conducts a rough 2D search (code delay and Doppler frequency) of the signal in order to assess whether the signal is present or not. The channel remains in acquisition mode until a signal is detected and a first estimate of the code delay and Doppler frequency is computed. At that point, the channel goes into tracking mode where the estimates of the code delay and the Doppler frequency are continuously refined and the navigation message is decoded.

In tracking mode, the channel monitors the quality of the tracking results in order to assess if the signal is present and is being correctly tracked. If the lock detection indicators fall below a threshold, the signal is considered lost and the channel goes back to acquisition mode and re-starts the whole process. The signal can be lost due to several factors such as shadowing of the satellite, cycle slips on the signal, or simply due to noisy initial estimates provided by the acquisition mode.

Receiver Level

At receiver level, the outputs of the channels (measurements and navigation message) are processed into usable solutions. Depending on the application, a GNSS receiver may need measurements from a minimum of three satellites (e.g. for 2D positioning), four satellites (e.g. for 3D positioning) or more satellites (e.g. for liability critical applications). Therefore at global level, if the receiver has enough measurements to provide application data it outputs a solution, otherwise it remains in idle mode. Some receivers might cope with instantaneous lack of measurements by extrapolating from previous measurements or even using information from external sources to support this extrapolation. An example is the possibility to use measurements from an Inertial Measurement Unit (IMU) in order to continue estimating positioning, even in degraded mode. Furthermore, the receiver may also use measurements from a satellite with bad quality indicators just to provide a solution, even if degraded, e.g. in case availability is more important than accuracy for its specific application. This is often the case for receivers that provide 2D solutions when only measurements from three satellites are available, e.g. in car applications where it is reasonable to assume that the user positioning can be map matched to a given road and therefore availability is the key performance factor.

Advanced Operations

On top of the basic operations at channel and receiver level, the overall receiver can further benefit from information either collected from past operations or recently decoded. For example, the receiver can use the information conveyed in the navigation message of a given satellite – e.g. almanacs from other satellites – in order to estimate which satellites are visible and therefore assign channels to those satellites that have a higher probability to be in view. Following these principles, acquisition is often sub-categorized in: cold, warm and hot start mode.

  • In cold start, the receiver has no prior information about its own position or which satellites are in view. As a consequence, each channel has to perform extensive search of all possible signals / satellites over all possible code delay / Doppler frequency pairs. It is only when (at least) one full navigation message is decoded that the receiver can infer the remaining satellites in view (through almanac information), and may redirect the other channels to follow them.
  • In warm start, the receiver has access to rough initial position (e.g. last position before powering down the receiver), and to almanacs relatively up to date (e.g. via external sources or stored at power down). With this information, the receiver is able to predict which satellites are most likely to be in view, and estimate their rough code delay and Doppler frequency, hence being able to narrow down the acquisition search space.
  • In hot start, the receiver has access not only to its rough initial position, but also to the ephemerides of the satellites, hence greatly reducing the acquisition search space, improving the time to acquire and therefore the final Time To First Fix (TTFF).

Related articles

References

  1. ^ Borre, K. et al, "A Software-Defined GPS and Galileo Receiver - A Single-Frequency Approach", Birkhäuser, chapter 5.